Dataset Viewer
Auto-converted to Parquet
query-id
stringlengths
2
5
corpus-id
stringlengths
1
5
score
stringclasses
1 value
11546
18572
1
82440
23010
1
82440
77782
1
82440
37697
1
82440
76985
1
82440
82489
1
82440
40412
1
82440
81115
1
82440
66489
1
82440
78789
1
36706
38722
1
36706
75291
1
36706
37292
1
36706
64505
1
19971
17152
1
19971
69036
1
19971
19894
1
19971
13405
1
19971
24575
1
19971
61376
1
19971
27749
1
19971
43042
1
19971
58123
1
19971
60680
1
19971
14766
1
19971
24966
1
19971
34078
1
19971
8616
1
19971
57628
1
19971
64304
1
19971
81224
1
19971
79509
1
19971
664
1
19971
78309
1
19971
51579
1
19971
26726
1
19971
6840
1
19971
25492
1
19971
46010
1
19971
78842
1
19971
22484
1
19971
55451
1
19971
10069
1
19971
58488
1
19971
11120
1
19971
15895
1
19971
69301
1
19971
4184
1
19971
49594
1
19971
25648
1
19971
36469
1
19971
55888
1
19971
768
1
19971
26161
1
19971
5094
1
19971
57171
1
19971
10990
1
19971
20113
1
19971
48214
1
19971
5443
1
19971
34361
1
19971
38551
1
19971
70236
1
19971
25258
1
19971
73035
1
19971
65709
1
19971
56822
1
19971
53675
1
19971
32584
1
19971
37220
1
19971
60863
1
19971
181
1
19971
44632
1
19971
23808
1
19971
81559
1
19971
77256
1
19971
64375
1
19971
24604
1
19971
40
1
19971
37532
1
19971
37314
1
19971
23590
1
19971
58250
1
19971
35014
1
19971
25552
1
19971
71339
1
19971
81967
1
19971
8037
1
19971
25164
1
19971
67221
1
19971
77077
1
19971
34136
1
19971
23515
1
19971
47027
1
19971
75791
1
19971
23349
1
19971
42377
1
19971
15715
1
19971
15779
1
19971
69357
1
End of preview. Expand in Data Studio

Dataset Summary

CQADupstack-android-Fa is a Persian (Farsi) dataset developed for the Retrieval task, with a focus on duplicate question detection in community question-answering (CQA) forums. This dataset is a translated version of the "android" (Android Enthusiasts) StackExchange subforum from the English CQADupstack collection and is part of the FaMTEB benchmark under the BEIR-Fa suite.

  • Language(s): Persian (Farsi)
  • Task(s): Retrieval (Duplicate Question Retrieval)
  • Source: Translated from English using Google Translate
  • Part of FaMTEB: Yes — under BEIR-Fa

Supported Tasks and Leaderboards

This dataset evaluates text embedding models on their ability to retrieve semantically equivalent or duplicate questions in the Android usage and development domain. Results can be viewed on the Persian MTEB Leaderboard (filter by language: Persian).

Construction

The dataset was created by:

  • Translating the "android" (Android Enthusiasts) subforum from the English CQADupstack dataset using the Google Translate API
  • Preserving the CQA structure for evaluating duplicate retrieval tasks

According to the FaMTEB paper, the BEIR-Fa datasets underwent translation quality evaluation using:

  • BM25 retrieval performance comparisons (English vs Persian)
  • GEMBA-DA evaluation, leveraging LLMs for quality control

Data Splits

As per the FaMTEB paper (Table 5), all CQADupstack-Fa sub-datasets share a pooled evaluation set:

  • Train: 0 samples
  • Dev: 0 samples
  • Test: 480,902 samples (aggregate)

The cqadupstack-android-fa subset contains approximately 25.4k examples (user-provided figure). For more granular splits, consult the dataset provider or FaMTEB documentation.

Downloads last month
77